AI readiness: when a shopping agent visits, can it buy?
Agents read a store as text and structure, not as a screen. 12 rules decide whether yours is readable, holdable and buyable.
Runs a real purchase. Costs us money. Please use your real store. We check it is a store first, in a few seconds.
Coverage 74 of 115 rules judged
Real audit, unedited. Numbers count up as the page loads.
Recent stores, real scores.
| Store | Score | Weakest category | Highest severity finding |
|---|---|---|---|
| wpfactory.com | 70 | Trust signals 38 | CHK-001Guest checkout offered as first-class option (no account wall) |
| enviafloresparadifuntos.com | 56 | Trust signals 30 | CRT-001Cart shows full cost breakdown: items, shipping estimate, tax estimate |
| tentree.com | 58 | Trust signals 31 | CHK-016Errors preserve entered data (never clear the form on error) |
| iliabeauty.com | 62 | Trust signals 34 | CHK-016Errors preserve entered data (never clear the form on error) |
| authenticpiecesjo.com | 52 | Trust signals 28 | CRT-001Cart shows full cost breakdown: items, shipping estimate, tax estimate |
Median 58 of 100 across 5 published audits. Updates as stores are published. All benchmarks.
Can ChatGPT buy from you?
Scored 0 to 100, separate from your conversion score. Framed as an opportunity, never as a defect.
We do not read your code. We buy something.
AI shopping agents read a store as text and structure, not as a screen. If price and stock exist only in scripts or images, if the cart has no stable address, or if checkout hides behind a CAPTCHA with no other path, the agent leaves. We score 12 such rules as a separate AI Readiness Index from 0 to 100.
Most stores we have published score between 52 and 70 out of 100. The median is 58, across 5 published audits. The two weakest areas are Trust signals, with a mean of 32.2, and Cost transparency, at 41.4.
The index is reported next to the conversion score, never inside it, and framed as an opportunity: the cheapest fixes, product schema and prices as plain text, are the ones that matter most.
All 12 agent rules: four on the product page, three on the cart, five on checkout.
115 rules in total. 7 critical, 31 high, 62 medium, 15 low. Every rule is public: the method, all 115 rules.
Complete Product schema.org: price, availability, GTIN/SKU, rating, shipping details
The structured product data that AI shopping agents read (ChatGPT, Google, Copilot) is incomplete on this page. Some or all of these are missing: product ID (GTIN), brand, review data, availability, shipping details.
AI-referred traffic converts 42% higher than regular traffic (Adobe, 2026). But agents only recommend stores whose product data they can read. Incomplete data means invisible to AI shoppers.
From the wpfactory.com audit, 25 Aug 2026, score 70. Every failing rule comes with this.
Or $49 a month for unlimited stores. 14 day refund, no questions.
200 orders a month. One critical fix worth 2 to 5% of checkouts. 4 to 10 more orders, every month, for $49.
example, using the estimate range the report printsDoes a low index hurt my conversion score?
No. The AI Readiness Index is reported separately and never lowers the conversion score.
Do I need an llms.txt file?
Google says it does not affect Search. Agents can still use it, and it is cheap. Rule PDP-A03 checks for it.
What are ACP and UCP?
Protocols that let an agent complete a checkout. Rule CHK-A01 checks whether a store declares one.
Is this only for large stores?
No. Product schema and prices as plain text are small changes any store can make.
Is it really free?
The score and three findings are free. The AI Readiness section is part of the full report.
Is this an AI guessing?
No. 115 fixed rules, each judged pass, partial or fail, each with a screenshot as proof.
See what a shopping agent sees.
About three minutes. Free.